How can a few names express a real prediction?

1 · Observe

A listing quotes square feet; your model wants square metres

A flat is advertised at 1,075 sq ft for ₹62.5 lakh. Your model was trained on square metres. If you feed it the raw number, it will happily produce a nonsense price, because Python does not know what the numbers mean. The first skill is knowing what each value is, what it is called, and what it measures.

2 · Express

Types, names and operators

Every value has a type: int for whole numbers, float for decimals, str for text, bool for True or False. A name such as price points at a value, and rebinding the name does not change the old value. Operators combine values, and the type of the result depends on the operator.

3 · Visualize

What the listing gives

The last row is the silent bug: dividing by the wrong unit gives a number ten times too big, and nothing crashes.

4 · Discover

Floats and names

0.1 + 0.2 == 0.3 is False, because floats store binary approximations. Compare floats with a tolerance. And after backup = price, followed by price = price * 1.05, backup still holds 62.5 while price is 65.625: assignment points a name at a value, it does not link two names.

5 · Apply

Two worked examples

Example 1, unit conversion. 1,075 × 0.092903 = 99.87 sq m. Keep the unit in the name: area_sqm, not area.

Example 2, integer division. Splitting 7 flats between 2 agents: 7 // 2 = 3 each and 7 % 2 = 1 left over.

6 · Where it goes wrong

Three ways it goes wrong

7 · In Python

Values and names in code

Here is how to implement it, step by step. The code below does exactly these steps, and every line is editable.

1. Write each value in a named variable, with the unit in the name (area_sqft).
2. Convert units explicitly: sq m = sq ft × 0.092903.
3. Compute price per sq ft from the matching unit.
4. Compare floats with a tolerance, not ==.
5. Remember assignment points a name at a value: backup = price does not link them.

area_sqft = 1075            # a listing quotes square feet
area_sqm = area_sqft * 0.092903
price_lakh = 62.5
rate = price_lakh * 100000 / area_sqft
print("type of each name:", type(area_sqft).__name__, type(area_sqm).__name__, type(price_lakh).__name__, type(True).__name__, type("Pune").__name__)
print("area in sq m:", round(area_sqm, 2), "| price per sq ft in rupees:", round(rate, 1))

# division, floor division and remainder
print("7 / 2 =", 7 / 2, "| 7 // 2 =", 7 // 2, "| 7 % 2 =", 7 % 2, "| 2 ** 10 =", 2 ** 10)

# floats are approximate
print("0.1 + 0.2 == 0.3 ->", 0.1 + 0.2 == 0.3, "| equal within a tolerance of 1e-9:", abs((0.1 + 0.2) - 0.3) < 1e-9)

# names point at values; rebinding does not change the old value
price = 62.5
backup = price
price = price * 1.05
print("price", round(price, 3), "| backup", backup)

# mixing units is a silent bug
print("rupees per sq ft if someone divides by sq m by mistake:", round(price_lakh * 100000 / area_sqm, 1))

type of each name: int float float bool str
area in sq m: 99.87 | price per sq ft in rupees: 5814.0
7 / 2 = 3.5 | 7 // 2 = 3 | 7 % 2 = 1 | 2 ** 10 = 1024
0.1 + 0.2 == 0.3 -> False | equal within a tolerance of 1e-9: True
price 65.625 | backup 62.5
rupees per sq ft if someone divides by sq m by mistake: 62580.9